November 2006 - Wine-devel - list.winehq.org

Re: wined3d: Respect ARB_VERTEX_BUFFER_OBJECT in loadVertexData.
by Markus Amsler 04 Nov '06

04 Nov '06

Stefan Dösinger wrote: > Am Samstag 28 Oktober 2006 16:44 schrieben Sie: > >> Its a Radeon Mobility M7 LW, with direct rendering. It has a RV200 core >> which is a R100 with the memory controller from R200. And it looks like >> the R100 doesn't support vertex_buffer_objects (couldn't find precise >> specs). >> > How well does bf1942 work without vertex buffers? My experiance with my M9 is > that it got a 100% performance boost with vbos. > It's playable (~10-15fps), perhaps 1/3-1/5 of w2k performance. > That were rather the vertex fixups, not the vbos themselves. I think you can > get most of that gain by improving the vertex buffer code to be able to do > vertex fixups without VBOs > Thanks for the tip. I will have a look at it.

2 2

Re: WineD3D: Scissor rect corrections
by Alexandre Julliard 04 Nov '06

04 Nov '06

Stefan Dösinger <stefan(a)codeweavers.com> writes: > + GetWindowRect(((IWineD3DSwapChainImpl *)This->swapchains[0])->win_handle, &windowRect); > + /* Warning: glScissor uses window coordinates, not viewport coordinates, so our viewport correction does not apply > + * Warning2: Even in windowed mode the coords are relative to the window, not the screen > + */ Wouldn't GetClientRect be more appropriate? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: secur32: [4/3]Test and fix DecryptMessage for multiple data buffers.
by Dmitry Timoshkov 04 Nov '06

04 Nov '06

"Kai Blin" <kai.blin(a)gmail.com> wrote: > --- a/dlls/secur32/ntlm.c > +++ b/dlls/secur32/ntlm.c > @@ -1345,22 +1345,32 @@ static SECURITY_STATUS SEC_ENTRY ntlm_Ve > if(helper->neg_flags & NTLMSSP_NEGOTIATE_SIGN) > { > SecBufferDesc local_desc; > - SecBuffer local_buff[2]; > + SecBuffer local_buff[pMessage->cBuffers]; This construct is not portable, you need to use HeapAlloc to dynamically allocate buffers. -- Dmitry.

1 0

Please commit my lostwages patch
by Scott Ritchie 03 Nov '06

03 Nov '06

Hey, I'm not sure who is responsible for it, but could someone please get my lostwages patch committed today? I've been getting a bunch of emails about Ubuntu Edgy because the website hasn't been updated yet. Thanks, Scott Ritchie

1 0

Tracking memory allocation
by Jeff L 03 Nov '06

03 Nov '06

I am looking at how to track whether memory has been allocated or freed. The SCRIPT_STRING_ANALYSIS opaque point in usp10 can be freed in a number of places. It seems that Windows handles this by checking a list of allocated storage as the passed value of a SCRIPT_STRING_ANALYSIS could be invalid. An obvious problem is where SCRIPT_STRING_ANALYSIS pointer is passed or copied to another pointer. At best one pointer could be NULLed but not both. I have seen this behaviour. It seems that this would not be a unique problem in wine, is there a standard way to handle memory tracking so that if memory is freed, at a later time we can check to see if the pointer is still valid? Jeff Latimer

2 7

oleaut32: fix regression at MSFT typelib parsing
by a_villacis＠palosanto.com 03 Nov '06

03 Nov '06

Finally, I managed to figure out the proper fix for the crash on regsvr32 msvbvm60.dll (and probably some other DLLs too). It turns out that the field from which TLBFuncDesc->Entry is filled is supposed to indicate the function entry point that implements the function described in the typelib. The roper interpretation of the entry is right there as a comment: BSTR Entry; /* if its Hiword==0, it numeric; -1 is not present*/ What this means is that if TLBFuncDesc->Entry has a high word of 0, it is supposed to be a DLL function ordinal. Otherwise, it is an offset of the function name (for lookup-by-name in the DLL). When this field was directly copied from the binary block (and the block kept around, with a memory leak), everything worked as expected. However, the patch to allocate space for the function name in the Entry-as-funcname case broke the Entry-as-ordinal case, since the ordinal value is not a proper pointer, and therefore it is incorrect to use it as such (like calling SysAllocString on it). This patch fixes the regression by copying the ordinal value when indicated by the FKCCIC flag, and checking the HIWORD before trying to free the allocated string in the Entry-as-funcname case. BTW, could any of you please comment on the previous patch for removing the use of CreateCompatibleDC in the implementation of GetDIBits() ? I sent it a few days ago, but it seems to have been ignored (no commit, no critique on why it is being rejected). Otherwise I will send it again in the assumption that it slipped through the cracks somewhere. Changelog: * Fix regression on MSFT typelib parsing of function records by allocating a string copy only when indicated by FKCCIC flag, and preserving the meaning of value as function ordinal otherwise. Alex Villacís Lasso

2 2

Re: missing locale resources
by Kirill K. Smirnov 03 Nov '06

03 Nov '06

> If you are adding multiple strings then the easiest is to start a new > block. Otherwise if it's a single string, or it logically belongs to > an existing block, then yes you have to add it everywhere if you want > a fallback. Thanks for solution!

1 0

Re: Missing locale resources
by Kirill K. Smirnov 03 Nov '06

03 Nov '06

> Strings are stored by blocks of 16, so if others strings in the same > block are defined for Russian the resource loader won't fall back to > English even if individual strings are missing. So, what is the right way to add new string? 1) Add to every resource Bg,Cs,Da etc... untranslated to be sure that string will be loaded. 2) Add to En and Ru only. Then check return value of LoadString and if zero do LoadStringEx(...LANG_ENGLISH/SUBLANG_NEUTRAL) 3) other way? Thanks in advance -- Kirill K. Smirnov

3 2

winex11.drv: Bug#6501. XIM problem with XInitThreads().
by Byeong-Sik Jeon 03 Nov '06

03 Nov '06

http://www.winehq.org/pipermail/wine-cvs/2006-October/026680.html > winex11.drv: Give XInitThreads another chance. > > --- > > dlls/winex11.drv/x11drv_main.c | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/dlls/winex11.drv/x11drv_main.c b/dlls/winex11.drv/x11drv_main.c > index 0403b5d..95f9d10 100644 > --- a/dlls/winex11.drv/x11drv_main.c > +++ b/dlls/winex11.drv/x11drv_main.c > @@ -383,6 +383,8 @@ static BOOL process_attach(void) > > /* Open display */ > > + if (!XInitThreads()) ERR( "XInitThreads failed, trouble ahead\n" ); > + > if (!(display = XOpenDisplay( NULL ))) return FALSE; > Hi, This patch makes the Bug #6501. Also Korean XIM(nabi, scim) have the same problem. Please fix this problem.

3 3

Missing locale resources
by Kirill K. Smirnov 03 Nov '06

03 Nov '06

Hi! I want to add a new string resource to winhelp program. I've modified only English-localized .rc file. My locale is Ru. When I trying to load this string, LoadString() function returns 0 and GetLastError() returns 0 too. I expect English string to be loaded, but no string is loaded. When I explicitly set locale LC_ALL=C winhelp then string is loaded OK (nothing else expected). So, the question: is this behaviour of resource loader correct? Should it fallback to LANG_ENGLISH SUBLANG_DEFAULT if localized string not found and LANG_NEUTRAL/SUBLANG_NEUTRAL is requested (as LoadString does)? Thanks in advance -- Kirill K. Smirnov

2 1